gis and participatory approaches in natural resources research

SOCIO-ECONOMIC METHODOLOGIES

FOR NATURAL RESOURCES RESEARCH

BEST PRACTICE GUIDELINES

GIS AND PARTICIPATORY

APPROACHES IN NATURAL

RESOURCES RESEARCH

Julian Quan, Nicoliene Oudwater,

Judith Pender and Adrienne Martin

Natural Resources Institute

The University of Greenwich

Published by Natural Resources Institute

© The University of Greenwich 2001

The Natural Resources Institute (NRI) of the University of Greenwich is aninternationally recognized centre of expertise in research and consultancy in theenvironment and natural resources sector. The Institute carries out research anddevelopment and training to promote efficient management and use of renewablenatural resources in support of sustainable livelihoods.

Short extracts of material from this publication may be reproduced in any non-advertising, non-profit-making context provided that the source is acknowledged asfollows:

QUAN, J., OUDWATER, N., PENDER, J. and MARTIN, A. (2001) GIS andParticipatory Approaches in Natural Resources Research. Socio-economic

Methodologies for Natural Resources Research. Best Practice Guidelines. Chatham, UK:Natural Resources Institute.

Permission for commercial reproduction should be sought from the ManagingEditor, University of Greenwich, Central Avenue, Chatham Maritime, Kent ME44TB, United Kingdom.

Production of this publication was funded under project R6800, by the PolicyResearch Programme of the United Kingdom’s Department for InternationalDevelopment.

The Department for International Development can accept no responsibility for anyinformation provided or views expressed.

Copies of this publication can be obtained by writing to NRI Catalogue Services,CAB International, WALLINGFORD, Oxon OX10 8DE, UK. When ordering,please quote BPG9.

University of Greenwich, a registered charity and company limited by guarantee, registered in England

(Reg. No. 986729). Registered Office: Old Naval College, Park Row, Greenwich, London SE10 9LS.

Natural Resources Institute

ISBN 0 85954 496 – 6

INTRODUCTION

Geographical information systems (GIS) have an important role to play in naturalresources (NR) research to support rural livelihoods, in particular, and pro-poordevelopment more generally. In this Guide we do not attempt to address the wholerange of issues associated with applications of GIS in development, but addressspecific questions relating to:

• stakeholder and beneficiary participation in the data collection process;

• participation in the planning process: assessment of planning and managementoptions, conflicts and development scenarios;

• integration of social and natural science information using spatial databases innatural resources research and development.

A rationale is presented for the integration of GIS and participatory approaches,highlighting the needs for close interdisciplinary collaboration and the application ofGIS to development processes through interaction with local stakeholders. Theissues involved are illustrated by examples of participatory GIS applications, and bypractical case studies in Brazil, Tanzania and Uganda, and Ghana (found at the endof the Guide), and literature-based examples from South Africa. The basicrequirements for making effective use of GIS are discussed along with, in aninterdisciplinary, participatory context, the methodological issues involved in datacollection, integration of biophysical and socio-economic data, data management,and data feedback and availability to stakeholders. Areas for further research anddevelopment are considered and overall recommendations on best practice are made. A glossary of terms used in GIS and participatory approaches in NR research isprovided at the end of this Guide along with a list of further reading and a contactlist. Guidance is also provided on suitability and cost-effective choice of hardwareand software for GIS and participatory approaches.

GIS AND PARTICIPATORY GIS

What is GIS?

A geographical information system is a computer-based tool for mapping andanalysing spatially referenced data; the advantages and features of GIS are shown inBox 1. GIS can facilitate the understanding of spatial aspects of social and economicdevelopment by:

• relating socio-economic variables to natural resources and the physical world;

• providing a tool for targeting interventions and monitoring impacts at variousscales and over wide areas; and potentially

• putting planning and research technology into the public domain to enrich – andenhance access to – information, to promote discussion and improveunderstanding of conflicting viewpoints.

As personal computers (PCs) and other information and communicationtechnologies (ICTs) become more sophisticated, globalization of these computer-

GIS and Participatory Approaches1

based technologies promotes widespread availability, lower costs and newopportunities. Despite some scepticism about the sustainability of GIS technologytransfer to the developing world, GIS is becoming a more accessible tool for storingand analysing information, mapping, visualizing and modelling developmentscenarios, and for monitoring progress and change. GIS can be applied at variousscales and levels of complexity, and dedicated systems for application in a variety ofspecialized contexts could become as standard on the average computer asspreadsheets, word processors and databases.

What is participatory GIS?

Participatory GIS is the integration of local knowledge and stakeholders’perspectives in the GIS. Stakeholders should also have access to GIS databases andproducts and be able to apply GIS and GIS products to development planning,resource management and advocacy. A variety of terms and acronyms are used bypractitioners, such as participatory GIS (P-GIS), which is used throughout thisGuide, GIS with participation (GIS-P) and community or stakeholder integrated GIS– but essentially these all refer to the same set of concepts and practices.

In considering participation, it is important to be aware of the distinctions andlinkages between primary stakeholders (the ultimate beneficiaries, i.e. localcommunities and the poor) and secondary stakeholders (institutions involved in thedelivery of assistance) in processes and projects which involve GIS. The direct usersof GIS and its products – maps, forecasts, tables and conclusions about developmentoptions and scenarios – will generally be government institutions but also,increasingly, NGOs and community-based organizations. In the development of P-

GIS and Participatory Approaches 2

BOX 1: Geographical information systems – advantages and features of

interest

A geographical information system (GIS) is a computer-based tool for mapping and analysingspatially referenced data, i.e. data related to defined geographical space. GIS allows integration ofdisparate layers of information into a common spatial database, whether the data are related to anexplicit geographical reference (e.g. latitude/longitude, grid) or to an implicit reference (e.g. a placename, road, environmental feature).

Technical advantages in data management

• Integrating common database operations such as query and statistical analysis with mapping,visualization and spatial analysis

• Integrating information from different sources and scales• Allowing management of complex information and making it more accessible• Ability to query with standard database management system tools (DBMS) as well as spatial tools• Spatial interpretation, predictive modelling and strategy planning• Rapid visualization, either on screen or as hard copy • Ability to update spatial data, e.g. derived from recent aerial photographs, satellite imagery or

ground survey

Features of interest for socio-economic analysis and social development

• Spatial perspectives with overlays of different types of data• Facilitating the management of large datasets and the integration of different datasets• Potential to incorporate stakeholder communication, debate, identification of conflicts; and

trade-offs in land and resource use; a tool in developing consensual approaches to planning• Empowerment of disadvantaged stakeholders with access to data and use as an advocacy and

planning tool

GIS, it is primarily these intermediary ‘bodies’ that need to involve primarystakeholders in the development process, and to deploy GIS technology to help meettheir needs.

There are potential benefits in developing participatory applications of GIS in thecontext of human-centred development models, such as the adoption of thelivelihoods approach within DFID, and by other development agencies and donorcommitments, to eradicate poverty.

GIS in rural development

The principal applications of GIS in rural development are community land andresource mapping, the integration of local and scientific spatial knowledge,community-based natural resource management (CBNRM), area planning andenvironmental management and the management of pests and natural hazards. Theapplications may be more, or less, participatory according to the data collection andanalysis techniques, the degree of stakeholder consultation and feedback and thelevel at which any management decisions are taken. Table 1 provides examples of thevarious participatory applications of GIS in NR research and development.

Risks associated with use of GIS

Since GIS has generally been an expert-driven technology, which is controlledcentrally by state agencies, scientific research institutions and private corporations, anumber of risks have been associated with its deployment in the service of human-centred development. These include:

• Only expert knowledge or data that are readily available in digital form – asopposed to local knowledge – will be incorporated in GIS.

• Planning decisions will be made by experts and technocrats with access to GIStechnology but without reference to those directly affected.

• Personal and community security may be violated if information supplied bylocal people is used by state authorities and developers without their knowledge,consent or understanding.

• GIS is relatively high cost and, unless safeguards are built in to ensure effectiveuse, the costs are unlikely to be matched by real social benefits.

These risks are real, but they can be addressed by deploying GIS in institutional andpolicy contexts in which there is a real commitment to incorporating the needs andperspectives of local people in development research and in planning and resourcemanagement processes.

FEATURES OF GIS AND REQUIREMENTS

Types of information handled

Geographical information systems can be used to organize information forappropriate delivery at different scales and, therefore, it is sensible to consider theneeds of users at different levels (national, regional, district and local) and to


GIS and Participatory A

pproaches4

Local level NR system

Community land and resource mapping;

land delineation and advocacy

e.g. Mozambique community landdelineation and registration project

Understanding and conservation of

indigenous knowledge/ integration with

scientific knowledge

e.g. Uganda/Tanzania soils research project(see Case Study 2 and Box 4)

Community-based NR management

e.g. Northern Namibia EnvironmentManagement

Environment management/regional

planning

Coastal environmental management e.g., Bahia Brazil (see Case Study 1)

Peri-urban environment management e.g. Kumasi NR Management Project (see Case Study 3)

Participatory methods

• Participatory mapping

• Use of GPS with local land users

• Listing, sorting and focus groups toexplore farmers’ soils categories

• Transect walks with GPS, householdinterviews, participatory mapping

• As above plus:

• Finer grained enquiry into localknowledge, resource use, seasonality,stakeholders’ interests; tenure and accessrights etc.

• Participatory planning methods

• Participatory socio-economic survey

• Participatory mapping and zoning

• Stakeholder iteration of maps andproposals

• Stakeholder workshops and analysis

Role of GIS

• Reconciliation of perceptual andconventional maps

• Digitization of participatory mapinformation and aerial photographs

• Provision of base maps and digitalimagery

• Production and updating of communityland resource maps

As above plus:

• Integration of scientific land resourcedata with local datasets; visualization and mapping

• Overlay, comparison and analysis of geo-referenced transect walks and mapsproduced by farmers and scientists

• Use of GIS maps derived from highresolution aerial photos and localknowledge

As above plus:

• Overlays of different stakeholders’perspectives

• Modelling and visualization ofmanagement scenarios

• Mapping of land cover and land use fromaerial imagery

• Production of base maps

• Integration of data layers

• Inclusion of social data and stakeholders’information

Level of stakeholder involvement(HIGH/LOW)

• LOW for mapping only – may beconfined to provision of local land andresource information

• HIGHER when maps are returned tocommunities and research is linked toadvocacy and CBNRM

• LOW but with extensive provision oflocal knowledge

• HIGHER if linked to farmers’participation in technology developmentand land management

• HIGHER if feedback to farmers whatdata/information are put into the system

• Potentially HIGH but may be uneven

• Risks of inaccessibility of GIS productsto local users and domination by localelite – active facilitation needed for mapinterpretation

• May be LOW, MEDIUM or HIGHaccording to attitudes of practitionersand institutional arrangements. Typicalvariables include:

• Commitments of state agencies to multi-stakeholder participation

TABLE 1: GIS applications in natural resource research and development

GIS and Participatory A

pproaches5

Watershed management e.g. Tamil Nadu Catchment ForestManagement (FAO, India)

Hazard management

Crop protection/pest management Red Seadesert locust control

Livestock disease control – cattle managementin tsetse-affected areas in Zimbabwe andTanzania

Forest fire management in Nicaragua andIndonesia

Limited:

• At regional scale generally confined tolocal agent reports on hazard sightings andoutbreaks

• Research at local scale may involvereconnaissance of local resource utilization,e.g. key informant interviews andparticipatory mapping, e.g. of grazing area,cattle density and migration

• Analysis of spatial data

• Stakeholders’ communication:visualization and mapping of socio-environmental characteristics

• Mapping of overlapping stakeholders’interests

• Modelling of planning scenarios –potential tool for conflict reduction andconsensus building

Fundamental:

• Modelling, monitoring and predictinghazard events and characteristics, e.g. pestpopulation dynamics, fire outbreaks andcontrol scenarios according toenvironmental conditions

• Uneven stakeholders’ access to andunderstanding of role of GIS

• Level of effective use of GIS withdevelopment and planning processes

• Level of data sharing amongst agenciesinvolved

• Institutional commitment and inter-agency planning meetings importantespecially at regional scale

• Local stakeholders’ involvement generallyLOW – confined to provision ofmonitoring information and localknowledge

• Demonstrable local benefits andinvolvement in hazard control requiredfor local participation

NR = natural resource, GIS = geographical information systems, GPS = global positioning systems, CBNRM = community-based natural resource management.

consider how data of different types and sources at various different scales can bemanaged. To be fully effective, GIS needs to link and map data derived from localknowledge, empirical surveys, topographical maps and satellite imagery or aerialphotography at suitable scales for micro-level case studies (e.g. 1:5000 or 1:10 000 scalemaps) and for wider overviews (1:50 000 upwards). Not all data will be susceptible tomapping. The integration of qualitative data into a common database, for example, bydisplaying a short text report or a scanned participatory sketch map via a hot-link,may ease its management and retrieval and set it in a geographical context.

Geographical information systems are useful as a tool for combining social andbiophysical datasets to facilitate and allow an integrated multidisciplinary analyticalapproach to an holistic understanding of particular natural resource managementquestions. For participatory applications of GIS in any sort of planning managementprocess, the crucial link is a spatial reference for all data entered into the system, i.e.to be from a specific place or derived from a particular physical area. An accuratespatial reference or geo-reference not only enables us to know exactly where we arebut allows information about particular places and areas to be displayed, analysedand used alongside other geo-referenced data, including biophysical datasets.

GIS technologies – software and hardware

Considerations in choosing the softwareTwo types of GIS can be distinguished based on the way data are stored andrepresented – vector and raster systems. A vector data model represents space aspoints, lines or polygons that are geographically referenced. The raster model dividesa space into equal sized cells (pixels) with attributes recorded as a numeric value foreach pixel; remotely sensed data are an example of raster data. Vector and raster datamodels fulfil the analytical and representational requirements of a GIS but there arebasic differences that should be considered when choosing a GIS. Both systems haveadvantages and disadvantages, not only in the type of analysis and modelling that isavailable in each concept, but also in practical terms such as data storage andcomputing time. Until recently, only a few top-end GIS software packages allowedboth data model types to be used together efficiently but, increasingly, thispossibility is now becoming available for PC and laptop use.

There is a wide range of GIS software available commercially, the capabilities andfeatures of which change rapidly – as do the hardware requirements. However, thechoice of software should be made according to the data being used and the outputsrequired. For example, if a project is concerned largely with remotely sensed data, araster system would be more appropriate; if the source data are primarily based ontraditional maps, and a high graphical output is required, a vector system should bechosen. Where local knowledge, participatory maps and participatory rural appraisal(PRA) findings are to be used with the GIS, and where maps and images are to beproduced for stakeholder feedback and discussion, a vector system should also bechosen.

When analysis and visualization of remotely sensed information and traditional orparticipatory vector-based maps are required, software that integrates both should beconsidered. A number of packages, such as ARCVIEW, are now available relativelycheaply and, given the advantage of capability to use remote sensing data, generallyprovide the best software for P-GIS applications.


Considerations in choosing the hardwareIn choosing computer hardware, the three most important factors for running asuccessful GIS are memory, speed and storage. Hardware to input the data (digitizer,scanner etc.) as well as to output the data (printers, data exchange), and provideimages and maps of sufficient size and scale should be an integral part of the system.Due to the visual nature of GIS outputs, the choice and availability of computerperipherals as well as the physical ambience and conditions in which the system ishoused are important. In areas of uncertain or erratic electricity supply, back-upgenerators or an adequate uninterruptible power supply (UPS) system should beprovided. Like the GIS software packages, the cost of computing hardware is fallingas standards rise.

The availability of a functioning global positioning system (GPS) and trained users isgenerally crucial during the research process, in order to obtain accurate geo-references rapidly for any type of field information. GPS is now relatively cheap andshould be used as essential ancillary equipment for any P-GIS application, especiallywhere field level data are required to supplement and update secondary existing datasources. In some cases, where adequate data or maps are available in geo-referencedform, such as aerial photography or high-resolution satellite images, and all featuresof interest are readily interpretable by the primary stakeholders, a GPS may not beneeded. For a fuller technical description of GPS see the Glossary of Terms.

Further guidance on GIS hardware and software and the likely costs can be found atthe end of this Guide.

Operating requirements

It is necessary to consider the principal tasks, skills and training requirementsinvolved in using GIS.

Stages in setting up and use of GISThere are several stages that can be recognized in the setting up and use of GIS.

• Feasibility study – the use of GIS in a project needs to be planned carefully. Skillsneeded to build a user-friendly bespoke package for users of only the outputs ofthe system are different from those needed to put together datasets for use in astandard software package. In order to rationalize the capture and storage ofinformation, this task should be given maximum importance and should includeparticipation of all the stakeholders – full consultation will benefit the next steps.

• Planning and design – an implementation plan, including the system and databasedesign, is needed – again with the full participation of all the stakeholders. Theroles of the project team should be defined and the availability of an adequateskill base to implement the plan in a workable time frame should be checked.

• Development – software and hardware are identified and purchased (see guidanceat the end of this Guide), databases are constructed and training undertaken.Operating protocols are also developed, especially for a large project.

• Implementation – the system is installed, data acquired and assimilated, andoutputs produced. Installation should be in a suitable working environment,


taking into account adequate space, good lighting and, if necessary, air-conditioning. Adequate precautions to protect the electricity supply should betaken (see above).

- Data acquisition can be considered in two main areas: background data, such asmaps and biophysical data which may need to be digitized, and project-generateddata. Simple measures, such as providing input forms for databases, standardizesand regularizes inputs – and checking and regular backup of data should beroutine.

- In a GIS that has defined output needs, the means of analysis can also bestandardized and user guides provided for the more casual users. This mayrequire the GIS practitioner to have programming skills.

- Data presentation, often in the form of maps, charts and tables, is produced toa high quality by plotters and printers. If hard copy outputs are required forfieldwork, lamination is advisable for protection and providing a more durablesurface for working on.

• Follow up and feedback – mechanisms are put in place for the system to beupdated and maintained locally if required. The initial impact of GIS technologycan be high – presentation of analyses is visually attractive. However, GIStechnology moves fast and if a long-term role is envisaged in a community orinstitution, the sustainability of the system must be considered. For example,help must be in place for dealing with possible hardware and software problems,and a means of upgrading the system must be identified.

Skills and training requirementsA range of skills and training requirements are required.

• Data collection and input – although sophisticated GIS skills are not required,tasks such as geo-referencing of field data and digitizing existing maps or aerialphotographs can be quite time-consuming.

• Data outputs – general computer skills and an awareness of the capabilities of theGIS.

• GIS practitioner using standard software – an understanding of GIS data models,encompassing GIS design, database design, data capture methods, transformationtechniques to enable data to be analysed in a common co-ordinate and projectionsystem, interpretation and analysis of data, storage and backup procedures andprovision and design of outputs.

• GIS practitioner providing a tailored GIS – all the skills of the above practitionerplus an ability to programme the chosen package.

• PRA researcher – an appreciation of the capabilities and limitations of spatialinterpretation and analysis. A working knowledge of the GIS, including the bestway to collect and present data for inclusion in the GIS.


Institutional and sustainability issues

Geographical information systems are used in a variety of institutional contexts. Thebasic requirement for effective deployment of GIS in a developing country is aninstitution in which management commitment and resources are sufficient to ensurethat staff have adequate skills, interest and resources to operate and apply thetechnology. To avoid dependence on data and technology imported from donorcountries, the project requires an institutional and business environment in whichbaseline data and maps are available or can be purchased commercially. CombiningGIS with a participatory approach requires a willingness to share datasets amongstinstitutional partners and an openness to stakeholder involvement in thedevelopment and application of GIS.

Sustainability is a key issue for project design if project support is to lead to anyenduring capacity to operate and apply GIS (and see next section). Projects with apurely research focus are unlikely to lead to sustainable GIS capacity, therefore it isimportant for research projects to link effectively with projects and institutionswhich benefit from adequate GIS infrastructure and operating skills.

Case studies documented in these guidelines benefited from project support invarious ways:

• In a DFID-supported coastal environmental management project in north-easternBrazil (Case Study 1), the partner institution had its own GIS capacity, which theproject upgraded and expanded to cover the project area by installing newequipment, funding a local GIS technician, accessing aerial photography andsatellite data, and providing consultancy support. A comprehensive database andset of maps of the project area are being produced for ongoing use, and theproject's social development consultant worked with the technical planning teamto promote a participatory approach. The state planning agency has begun torecognize the validity of local sources of information, to share data with otheragencies, and to respond to stakeholder needs and priorities in applying GIS in theplanning process.

• A research project combining scientific and indigenous knowledge of soils in EastAfrica (Case Study 2) relied upon GIS capacity with a European ‘bias’. Theemphasis was on methodological research using GIS rather than capacitydevelopment to make ongoing applications. This case illustrates the high level ofexpertise required to manage GIS databases successfully and use them as ananalytical tool. Data entry and database management was mainly at universitieswith GIS expertise in Belgium and UK, while data analysis and integration withindigenous knowledge involved a significant input from the Natural ResourcesInstitute (NRI) of the University of Greenwich. The developing countrycollaborators – agricultural universities, research station staff, local NGOs andextension officers in Tanzania and Uganda – had limited skills in data analysisusing GIS and training given had limited impact.

• The Kumasi GIS for peri-urban natural resource management (Case Study 3)involved parallel systems, run in the UK and Ghana, developed by NRI andinstalled using project funds with data managers in each country paid by theproject. In the final stages of a current project using the GIS, the host institute inGhana is being encouraged to develop a plan to sustain the system for local use


over the next two years. This will involve the development of GIS courses forlocal managers, planners and students, the commercial use of the data for localbusiness (e.g. in tourism planning) and a positive commitment to publicizing theuse of the GIS in future research.

PROJECT DESIGN, PLANNING AND PREPARATION

Importance of a multidisciplinary approach and participatory planning

The first consideration is to establish whether GIS is a relevant technique to use. WillGIS enable the production of new knowledge that can provide real benefits tostakeholders? In answering this question, it will be necessary to develop aninterdisciplinary approach from the outset, and to consider how the products of GISas a research tool will actually be used, the needs of different stakeholders and therole that different disciplines will play in the collection and analysis of data. If GISoffers relevant techniques, it does not make sense to design or commission projectsin which social and natural scientists and GIS practitioners develop lines of enquiryindependently of one another, seeking to integrate their findings and perspectivesonly at the end.

In assessing the relevance of GIS, it is critical to involve developing country partnersand collaborating institutions. The key criteria for determining relevance andapplicability of GIS in a given institutional context include:

• existence of institutional capacity to operate GIS• demand/need for the application of GIS to the research or development problem• presence of a data-sharing culture and possibilities amongst local agencies• availability of baseline cartographic data • at least some relevant thematic data in geo-referenced format• relevance of spatial analysis – typically, in understanding spatially differentiated

livelihood and production systems, area-based development and environmentalprocesses, and/or micro-meso-macro links

• innovation and demonstration value of conducting data analysis andcommunicating results using the capabilities of GIS.

Typically, but not exclusively, these conditions might be met in middle-incomecountries or those with research and planning agencies, focusing on the needs ofurban and peri-urban areas or more developed and populated rural areas. In projectdesign, consideration of institutional capacity and infrastructure amongst the usersis of major importance in choosing GIS software and hardware. Important issues totake into account are:

• financial – start up and running costs• available institutional infrastructure to collect the data, operate and apply GIS,

and to employ, train and retain suitable staff• personnel – level, range and number of skills, and opportunities for ongoing

training• awareness of the capabilities of a system – and provision of suitable

accommodation for it• commitment to a feasible programme of staff training and awareness raising of

GIS users


• data input – methods and means available• output – methods and quality.

In view of these requirements, it remains difficult to maintain and operate GIS atlocal or district level in many developing countries because of inadequateinfrastructure and skills. The more sophisticated the hardware and software thegreater the difficulties will be.

Selection of planning units and mechanisms for stakeholder involvement

Secondly, it is necessary to consider the specific objectives of the project, the scaleat which it is to operate, and how stakeholder/community level information andinvolvement are relevant. If GIS does promise to add value to a project, how shouldit best be used and at what level?

Where a project examines location-specific issues, it is likely that local knowledgeand participatory research and mapping will need to be conducted in order togenerate information on NR and livelihood systems. Feedback of outputs, probablyvia clear, simplified large-scale maps and thematic map overlays, is also useful tovalidate results with the information providers or to enable their participation inapplying the results.

Over wider areas, institutional participation and application of results becomerelevant in both research and technical co-operation (TC) contexts – and a P-GISapproach is especially relevant. It will be important to gather data at district orregional scale on the one hand and at local scale on the other, for instance, throughin-depth micro-level enquiries in a number of representative sites. Partner agencieswill need to be able to grasp the wider spatial relevance of micro-level knowledgeand, where they are engaged in area planning and management, will need to developlinks which enable participation of local stakeholders. The application of GIS mayhelp by producing maps to share the findings amongst stakeholders, and by enablingappreciation of the spatial significance of local processes and research results. Ifparticipatory planning methods are to be developed or tested, institutionalacceptance of stakeholder involvement and identification of opportunities todevelop this are essential features, without which GIS is likely to remain a centralizedand technocratic planning tool (see Case Study 1).

Once a project is initiated, researchers and project staff should consider the types ofdata needed, their likely uses, and the needs and opportunities for dialogue amongstthe users. In planning ahead, local information and participatory data should begiven equal weight with scientific information in developing the GIS, and surveymethods should be designed accordingly.

It should be remembered that despite problems of quality and compatibility of datafrom different sources, maps incorporating different sources and types ofinformation can create common frames of reference and facilitate dialogue. Maps arelikely to provide useful tools for planning a survey, relating the human population ofan area to its natural and physical features, and encouraging spatial thinking amongstinvestigators. An approach which proved useful in Brazil, as outlined in Case Study1, was to incorporate existing knowledge of the project area into sketch maps as avisual aid in selecting sample sites. More detailed, accurate maps can be prepared as


further data become available. A first task is likely to be the accurate spatialreferencing of survey sites, human settlements and the boundaries of the areas ofinterest.

SOCIO-ECONOMIC AND BIOPHYSICAL DATA COLLECTION

Tools for collection of socio-economic and other qualitative data

The type of data needed, and the methodology and sequencing of data collectiontechniques depend greatly on the type and objectives of a particular project and whatthe outcomes will be used for.1 The techniques for data collection suggested in Box2 offer a range of tools that can be selected according to the specific objectives andnature of the project. In general, the tools used in conventional PRA are appropriateto gather data for use with GIS. Careful attention needs to be paid, however, to thesequencing of different tools, how the information is entered into the GIS and howdifferent sources of information are linked together for analysis within the GIS.Spatial referencing of qualitative data should form a key component throughout thedata collection process. Importantly, both the PRA and the GIS developmentprocess should work towards the same overall purpose and be planned accordingly.Consistency in the composition of the research team is also important, since thisallows for regular cross-checks and feedback between the different data sources andteam members.

Socio-economic data sources

Pre-existing or published dataSecondary sources, i.e. pre-existing or published data, include PRA-derived datainformation and census information. Pre-existing socio-economic datasets may beextremely useful but in most cases they are not geo-referenced. Census informationis likely to have a cartographic reference related to map units but this may beavailable at too high a level of aggregation to assist with local enquiry. Alternatively,it may be based on census districts which do not correspond with administrative ormanagement units, such as local parishes, watersheds, agro-ecological zones ordesignated conservation areas (see Case Study 1).

Although pre-existing PRA data frequently relate to specific locations, they are alsogenerally not geo-referenced. Where systematic PRA has been undertaken overwider areas, using specific criteria to obtain representative coverage and capturesocio-economic and environmental diversity, it may be relatively easy to geo-reference and integrate with other datasets retrospectively. The use of key PRAfindings within or alongside a good map displaying other features of interest, such asinfrastructure and land use, may prove useful in developing a spatial understandingof, for example, the variation in livelihood opportunities and constraints across aproject area. This was done in P-GIS case studies in north-eastern Brazil and inKumasi, Ghana (Case Study 1 and 3).

Social and economic baseline and diagnostic surveysThese surveys are undertaken as components of the research or to support planningwork. In the design of participatory survey work, it is advisable to geo-reference


1 Examples of different types of projects are: research projects, land rights mapping, CBNRM, participation in landuse planning and devolved local planning/advocating alternative plans.

survey sites in advance or as the enquiry proceeds. The use of good maps – evensketch maps – will prove invaluable in identifying survey sites and in devisingsampling strategies which reflect important spatial variables, such as populationconcentration, agro-ecological features, and access to markets, services and labourmarkets. The accuracy and usefulness of the maps can be improved as work


BOX 2: Tools for quantative data collection

• Secondary socio-economic data, literature and project reports National census informationcompiled by national and regional governments, and research conducted by other organizationsand institutions can provide important background information. This sort of information canplay a key role in identifying areas to work in and in the selection of survey units. However, suchinformation is often outdated and complicated to link with other data due to non-spatialreferences and possible differences in scale and scope.

• Participatory mapping This is useful for exploring community members’ spatial conceptions oftheir natural and social resources, land boundaries etc. A widely applied tool in P-GIS due to itsspatial focus. However, there is a need for integrated use with other tools to put into context andunderstand local spatial perceptions. For linking with other databases within the GIS, it requiresa fair level of spatial accuracy. Perceptual mapping can be done in several steps, starting with afree-drawn map, followed by mapping based on photo maps or uniform base maps developedfrom aerial photographs and topographical maps, to allow for consistency and spatial accuracyfor use within the GIS. Close participatory observation and making notes about the actualmapping process is likely to help with the process of analysis.

• Semi-structured interviews These allow the participants more scope to investigate what peopleknow and to follow up topics of interest as they arise in the discussion, and can be used withgroups and individuals. A wide range or mix of criteria can be applied to select participants:gender, age, level of education, area of residence, socio-economic status, size and nature of landholding etc.

• Focus group interviews Group interviews provide exchanges between participants withdifferences of opinion that can often lead to greater insights into people’s perceptions. Care isrequired over the composition so that as many as possible feel free to express their opinions,especially those with less status who may be better interviewed in a separate group orindividually. They can take place at different stages during data collection, data entering and dataanalysis for cross-checking and feedback.

• Key informants interviews ‘Experts’ – those identified by local people as having specialistknowledge – may be interviewed, taking care that they are not confined to those with formaleducation and access to scientific and/or outsider knowledge.

• Field visits and transect walks These combine observation and discussion and are useful inallowing the participant/respondent to point things out in situ. They may also provide a morerelaxed atmosphere than a group meeting, making communication easier. Qualitative informationshould be geo-referenced with a GPS to allow spatial reference and overlays of thematic mapsduring analysis within GIS.

• Participatory observations These are useful for comparison of actual practice to the normspresented in group discussions or interviews. As with transects, qualitative data can be geo-referenced to support analysis at a later stage.

• Diagramming, ranking exercises and games These can be used to elicit local perceptions,definitions and classifications. Tools include ranking of importance, comparing characteristicsusing pair-wise ranking diagrams, seasonal calendars and network diagramming.

• Local classification systems/taxonomies This can be a difficult area as it requires anunderstanding of social and anthropological principles of research in combination with linguisticskills. Notion of multiple realities should be well understood. It involves the identification oflocal terms, then asking local people to sort and group the categories, identifying commonfeatures and contrasts in the context of the wider language and cultural system.

• Cultural expression The content of songs, poetry and speeches on celebrations and publicoccasions can reflect significant messages and social values.

Source: Adapted from Warburton and Martin (1999)

progresses using the GIS, for example, to capture previously unknown localsettlements and markets, and changing land uses.

Local and indigenous knowledge and perspectivesAnother source of socio-economic data is local and indigenous knowledge andperspectives derived from location-specific participatory enquiry, and includesmapping exercises and interviews. One of the main requirements for successfulapplication of P-GIS is the input of multiple sources of knowledge provided by andgenerated from the stakeholders involved, with a particular emphasis on localknowledge. Local people’s knowledge (LPK) includes the complex of practices anddecisions made by local people and is shaped by and interlinked with technical,cultural, political and social knowledge. It is based on experiences passed from onegeneration to the next but, nevertheless, it changes, adapts and assimilates new ideas;it can be quite location-specific and may vary among individuals according to age,gender, socio-economic status, ethnicity, area of residence etc.2 It is important toconsider the epistemological differences between local knowledge and so-calledscientific knowledge during the design and implementation of data collection andanalysis throughout the project cycle.

Biophysical data sources

Biophysical sources of data3 fall into three broad categories:

1. Cartographic and other published datasets2. Remotely sensed imagery – including aerial photography/videography and

satellite data3. Empirical survey data and ground-truthing (or on-site validation) of remotely

sensed data.

Secondary data sources, which include publications in digital form, can also include:

• topographic maps• digital elevation models• climatic and agro-meteorological data• soil survey information• geological information• vegetation and land cover maps• information about agro-ecological zones.

On-site validation and biophysical field survey may provide opportunities forinvolving local enumerators and key informants in generating local knowledge andinformation through participatory techniques, such as group discussions, perceptualmapping and transect walks using GPS. These techniques can be used to gatherethnobotanical information about resource utilization, environmental change, landaccess and ownership, user conflicts, relative local values and socio-economicimportance of different species. This approach is useful for validating, interpreting


2 For a more detailed discussion on local knowledge and participatory approaches see Warburton and Martin (1999).

3 Detailed guidance on the collection of primary or secondary biophysical datasets for GIS is not provided – nor isguidance on the use of remote sensing imagery and data available from published sources or commercial suppliers.

and updating, for example, aerial photography and remote sensing information, andunderstanding the socio-political and management implications. There are significantopportunities here for better integration of ecological and land use survey work intorural development planning process, and for developing fuller partnerships withlocal people, but these are frequently missed. The employment of local communitymembers as enumerators can also provide incentives for wider local participation insurvey work and management processes arising. (See Case Studies 1 and 3.)

Quality control of GIS data

The quality of GIS products is often judged by the final appearance of the printedoutput which can be visually appealing and can disguise poor quality of data. The useof inaccurate or unreliable data in GIS will, however, lead to defective maps andinaccurate analytical conclusions. In some instances, the only data that are availablehave to be used and it is, therefore, important that the quality of all data isacknowledged and understood by users of a GIS. Adequate procedures need to be inplace for deciding which data are worth using – and in what form. Some of thecommon sources of error are shown in Box 3.


BOX 3: Factors affecting the quality of GIS data

The quality of spatial data depends on:

• how recent the data are and how up to date• datasets in time series being comparable with datasets over the period of survey• completeness over an area• using different data layers that are comparable in scale• scale used being suitable for the study in hand• formats of different datasets being compatible• project having copyright or licence rights over the data being used• suitable datasets being available within the project budget• suitable density of field observations being made • locational accuracy of field observations being suitable for the scale used• attribute data being entered accurately and in a way that provides sensible analysis and outputs

In addition:

• topological accuracy must be maintained within and between spatial datasets• meta data are to be accurately entered and stored with the data• all inputting should be checked • choice of data model should be suitable for the project• account must be taken of imprecise and sometimes contested nature of boundaries • the risk of observer’s bias must be acknowledged • limitations of the computer processing and analysis must be understood

Errors specific to GIS may be generated by:

• problems in integrating and classifying data using map overlays based on datasets with differentformats and levels of accuracy

• inconsistent or imprecise use of logic in data interpretation and spatial analysis• error propagation and magnification during data transformation• choice of mathematical models and datasets in interpolating non-empirical information.

Source: Derived from Burrough and McDonnell (1998)

INTEGRATION AND ANALYSIS OF SOCIAL AND BIOPHYSICAL

DATA

Rationale for integrated research

Within applications of P-GIS, participatory and perceptual mapping is a widelypractised tool as it easily digitized and stored within the GIS. However, qualitativedata collected through participatory and perceptual mapping tend to be spatiallyinaccurate. This can be corrected by proper sequencing of the mapping process,starting with free-drawn maps which can then be reconciled with uniform base mapsdeveloped from aerial photographs and topographic maps (Mather et al., 1998).However, maps should not be used as a single tool to collect and explore localknowledge, since they do not allow for more complex concepts and interactions or‘triangulation (Chambers et al., 1989). An integrated approach using different toolsfor collection and analysis of both spatial and non-spatial data will enhance an in-depth understanding of locally produced perceptual maps and reflect differentstakeholders’ perspectives and realities of the same land area, for example, those offarmers, gatherers of forest produce, private developers, and local governmentplanners.

Uncritical rapid questioning and recording of qualitative data and local knowledgemay lead to misunderstanding, misconceptions and distorted results during theanalysis within the GIS until more detailed work can be done (see Box 4).

In addition, regular feedback with the local communities about the research processand preliminary results not only improves their understanding of local perceptions,but also gives them a sense of ownership of the process by the local communities. It


BOX 4: Exploring local soil categories in Uganda

The integration of different methodologies (e.g. mapping, transect walks, focus group discussions andindividual interviews, sorting tasks etc.) made it possible to reach an understanding of how farmers’expressions of their knowledge, and the soil categories used, relate to the particular location andsocial setting of the interview, discussion or observation and its sequence in the research process.Some categories were brought up in some exercises and not in others. Not surprisingly, mappingtasks appeared to elicit farmers’ definitions of land areas, while the sorting interviews and focusgroup discussions elicited nuances in characteristics of the actual soils, and the transect walksgenerated location-specific, detailed descriptions.

The term eitela is a good example. In Wera village, Uganda, the first attempt at participatory soilmapping using aerial photographs resulted in large areas of the village being designated as eitela (thisterm describes an upland area – often bushy and left uncultivated for a while – and is not a soil typeas such but a land use description that includes different soil types). During the sorting task, farmerswent into more detail as they grouped their different soil types, explaining the grouping criteria andthe similarities and differences between them, and produced a different understanding of what theconcept eitela actually means. Most farmers involved in the sorting task drew this distinction betweeneitela and soil types, and this was further clarified in the focus group discussion. A later map drawnby the farmers who had been involved throughout the research process did not include eitela.

Another example of the limitation of participatory mapping as a single tool was the exclusion oflocalized soil units, such as fields located on previous homesteads or anthills, as they are rather smallunits and are difficult to map. Although limited in coverage, these soil types play an important rolewithin the local agricultural production systems due to their high fertility.

Source: Derived from Case Study 2

allows them to have a certain level of control over the information entered and usedwithin the GIS throughout the process, thereby limiting potential conflicts andsecuring future community involvement (and see section on Feedback andAccessibility of Data).

Integrated approach to data collection and GIS analysis

Qualitative data are not readily suited to modelling and spatial analysis. However,qualitative enquiry can lead to the ranking or scoring of resource availability orquality, livelihood opportunities, socio-economic well-being, poverty indicators anddevelopment problems etc., across different areas and between different groups. Theresults can be visualized and mapped, and overlaid with available quantitative orsecondary data, such as population or health information, rainfall data, landholdingsize, and land cover information (see Box 5).

Biophysical data are often already available in a geo-referenced format as they aremainly obtained from remote sensing, aerial photography or cartographic sources,linked to a GIS. However, socio-economic and qualitative datasets are generally notgeo-referenced, and are therefore more difficult to integrate with quantitative setswithin the GIS and, as a result, are less susceptible to spatial modelling. Forsuccessful integration, there is a need for an integrated and iterative approach duringdata collection in which a GPS should play a significant role in gathering spatialreferences for qualitative datasets.

There are various ways in which qualitative information can be integrated andcombined with physical, environmental and other quantitative datasets within theGIS:

• entering qualitative data in a text database that can be linked with spatialreferences – this is especially important as a means of tracing whose knowledgehas been elicited, documented and incorporated into the GIS;4

• hot-linking spatial maps to descriptive texts;

• use of perceptual maps that are spatially referenced and linked to a qualitative textdatabase for in-depth information and clarification (e.g. in the case of fuzzyboundaries and multiple meanings of local perceptions – ‘good grazing land’,classifications of natural resources etc.);

• geo-referencing transect walks to allow for spatial reference of qualitativeinformation/local knowledge on, for example, natural resources management;

• entering qualitative data in structured databases containing tabulated, scored orranked qualitative information, and video and audio clips recording oral historiesand local people’s views and knowledge – this type of database can be plotted byspatial queries and multiple overlays using other thematic maps stored within theGIS.


4 Where personal or location-specific information is included in a database, data security, confidentiality andintellectual property rights issues may arise, especially if the information is to be published, networked or somehowsold on.

In some cases it is necessary to analyse the data before they are mapped, for example,by scoring and ranking the indicators or variables in question and assessing thespatial significance, coverage and variation in the data. This can be undertakenindependently and loaded into the database, or done by a spatial analyst using theGIS to undertake the analysis and reflect it in a mapped form.

Problems and challenges in integrated data collection and analysis

Data integration problems commonly arise as a result of failure to reconcile datasetswithin a common spatial and temporal framework, which allows for the differentialdata accuracy and reliability. For example:• Use of outdated biophysical or socio-economic datasets in multiple overlays with

more recent qualitative observations (e.g. derived from a village PRA) can resultin a distorted analysis and lead to biased recommendations. Both datasets need tobe validated by making qualitative observations over a wider area (scaling up thePRA, and ground-truthing the older datasets) and obtaining new survey data (e.g.from aerial photos or high resolution satellite images).

• In some cases, qualitative data can include multiple perceptions and fuzzyboundaries that are impossible or difficult to reference spatially. Despite theprima facie difficulties, this problem should be treated positively in bothinformational and planning terms. Technical possibilities of, for example,establishing hot-links to text databases can assist in linking the information to theGIS, but the key question is whether or not imprecise data and multipleperspectives can be actively used and incorporated within the analysis. Generally,users are tempted to reduce multiple perspectives into a single and easilyunderstood one. However, it is worth seeking to change this, since simplificationfor the purpose of computerized mapping may mislead users, leading toinappropriate interpretation and action.

• Imprecision of information (see Box 6) is no justification for omission. The useof fuzzy logic techniques enables handling of qualitative data rather than reducingthe data to restrictive quantitative classes within GIS. Fuzzy logic or fuzzy settheory allows individual pieces of data to be members of different overlappingsets – and uses possibility instead of probability as a statistical technique.


BOX 5: Exploring local soil categories in Uganda

Through participatory mapping, the community provided information on the location and type oftheir water sources, their specific uses (such as household consumption, watering cattle, washing andlaundry) and their state of repair. Hydrological surveys gave information on the chemical quality ofwater. Both databases were combined by overlaying thematic maps, and information on the qualityof water used for livestock and human consumption was obtained. Additionally, the number ofdamaged water points could also be identified more efficiently.

This provides a clear illustration of how a qualitative enquiry on resource availability and quality canbe mapped and visualized within the GIS.

Source: Cinderby, presentation at workshop Geographical Information Systems and Participatory Methods, NRI,Chatham, July 1999.

• Importance of cross-checking data with local communities before entering intothe GIS. In some cases data may need to be analysed before being mapped, forexample, scoring and ranking social well-being, natural resources quality,problems and needs, and key resources. This analysis can be done by the researchteam but the results should be presented to the local communities for feedbackbefore being entered and used within the GIS.

DATA STORAGE AND MANAGEMENT

It is necessary to match technical GIS development and operations sequentially tothe different tasks. At the outset, sketch maps or simplified maps produced usingGIS overlaying key features may be very important, even if their locations are notfully accurate. Subsequently, a project may need to obtain participatory sketch mapsfrom the field, topographical and thematic maps, and digital images, and reconcilethese to a consistent scale. The next stage is likely to involve entering into thedatabase the detailed empirical information, collected by the project, about resourceuse and socio-economic and biophysical features, and the production of maps torepresent this information spatially.

The most developed uses of the GIS will be as a data analysis tool and subsequentlya planning tool. Social and natural scientists will need to work closely with GISspecialists to determine the best approach to data analysis and the representation ofresults in maps and other graphical forms (e.g. combination of point-specific socio-economic datasets from particular villages alongside maps of resource utilization andphysical infrastructure). The quality of graphical and mapped outputs of GIS isparticularly important in relation to the need for feedback of data, as discussed inthe next section. As an analytical tool, GIS can also contribute significantly to thediscussion of spatial relationships and, for instance, to the evaluation of planningoptions in written outputs.

In developing and managing a GIS database:

• It is essential to build on existing data/information sources and existing GISsystems. Always seek existing digital sources of data since data entry is the mostexpensive and time-consuming element in GIS development.


BOX 6: Use of boundaries

If stakeholders perceive the boundaries on maps as accurate, rather than fuzzy, it may lead to conflictbetween different groups, for example, some communities used the lines they drew on maps to layclaim to communal land and to try to prevent other groups from entering ‘their’ subsistence area;this triggered local conflicts about access and ownership of land. In other cases, communities andother stakeholders decided not to define and include boundaries of communal/private land in orderto avoid such boundary disputes. Therefore, there is a need for capacity building to understand thelimits of any analysis within the GIS to avoid such conflicts emerging.

Source: Cinderby (1999)

• Collaboration in data exchange may lead to wider inter-agency andinterdisciplinary collaboration. Digital base datasets derived from the projectshould be made more widely available to a variety of end users.

• Meta data (i.e. data about available datasets and sources) should be documented,giving the source and ownership/copyright of the data and agreements relating totheir use. All in-country standards that exist on recording meta data should beadhered to. It is essential to track errors, omissions and imprecise features – forsubsequent improvement – so these can be included in the meta database.

• It is important to track changes in the data and processes applied to a dataset (filenames, projection changes, attribute joins, analyses). If there is no provision forthis in a meta database, a separate processing file should be kept.

• Secure backup systems should be in place – backups should be regular and, ifpossible, stored in a different location from the GIS.

• Most GIS can use data directly from spreadsheets and databases. Some GIS canhave direct access to statistical packages, such as ARCVIEW and S-Plus – S-plusalso has a spatial statistics package that links directly with ARCVIEW.

FEEDBACK AND ACCESSIBLITY OF DATA

Data display and feedback for participatory discussion

In order to communicate spatial information to stakeholders, good, clear accessiblemaps are essential. There are a number of basic principles:

• Do not rely on computers alone. Printed maps may be more versatile than GIScomputer files for permanent display, especially in remote locations. A notebookcomputer can be useful to manipulate and display data in an interactivediscussion, but images will need to be projected if viewed by more than a fewpeople.

• Maps should be produced in large format – A2 or A1. For discussion purposesand depending on complexity, A3 is probably the minimum size for practical useeven by researchers and technicians.

• Overlays should be clear, restricted to the most important variables and mainfeatures, eliminating ‘noise’ or irrelevant detail. Avoid providing too muchinformation in a single map. Different overlays can be produced to highlightdifferent features.

• Use of colour should be clear – avoid using shades of the same colour, indistinctjuxtaposition of colours, and complex shading or hatching to illustrate differentfeatures. Natural, realistic colour should be used where possible, for example,green for vegetative features, blue for aquatic, brown for degraded land or urbansettlements.

• Map legends and symbols should be clear, simple and always made available witha map.


• Clear simple graphics produced by a GIS are a useful adjunct to a map, or can beincluded in it if visually not too complicated.

Although there is concern and discussion about the intelligibility of maps to primarystakeholders who may not be visually literate, a growing number of people arecapable of interpreting a map – a basic skill which should be promoted. The use ofreadily recognizable images, such as aerial photographs, and the reconciliation oflocally produced maps with topographic maps and aerial photography or highresolution satellite imagery data, can aid interpretation. A common visualunderstanding, amongst local residents, researchers and planners, of the spatialfeatures of an area may also be a prerequisite for joint understanding andmanagement. Moreover, clarity and accessibility in map production will make themaccessible to all stakeholders. Too often, highly complex maps produced byspecialists using GIS are incomprehensible even to visually literate professionals.

Providing feedback and maps to local communities and information providers servesa number of useful purposes, including:

• validation of survey findings and maps produced• generation of additional information and understanding of stakeholder

perspectives through discussion• promotion of local involvement in a planning and management process• communication of information to decision-makers • encouragement of discussion of planning scenarios and resolution of potential

resource use conflicts.

Control of GIS technology and accessibility of products to local

communities

A key factor in the successful implementation of P-GIS lies within the partnershipscreated between all participants involved – spatial analysts, social scientists,government officers, NGOs and local groups. Availability of and access to spatialreferenced data/GIS can contribute to capacity building for environmentalmanagement at local level, and help empower local communities. Throughvisualization, GIS can be a useful tool to increase understanding of all participantsand foster constructive discussions (see Cinderby, 1999).

Although accessibility of GIS to the community or even local ownership and controlof the system may be desirable, the problems of operating GIS at local communityor district level – costs, lack of skills, and maintenance problems – mean that‘community integrated’ GIS is relatively untested and, in most cases, a long way off. The use of GIS may sometimes be useful in promoting institutional collaborationand responsiveness to primary stakeholder perspectives in otherwise closed,disabling institutional contexts, as occurred in the Brazil case study. However, asremarked at a P-GIS workshop, GIS, like PRA, is only as good as the institutionswhich use it, and its applications will only be as participatory as the local politicspermits (Robert Chambers, personal communication, 1988). If the planning processis not responsive and accessible to local people, GIS alone will not change thesituation.


Secondary stakeholder participation in GIS development can also strengthenownership or projects and project outputs: common interests in GIS and its dataproducts can provide incentives for institutional collaboration. However if agencies,involving or representing the poor and marginalized groups, are not involved inplanning and development, there are serious risks that GIS will remain a top-down,expert-driven system, which centralizes and does not divulge spatial information forpublic use.

Technical investment in GIS development and in data collection processes, howeverparticipatory this may be, is not enough. For P-GIS to be effective, theresponsiveness of institutions to social need is fundamental. Institutions developingGIS for development purposes need to reach out to local stakeholders and developnew opportunities and mechanisms to involve them in research and planningprocesses. This must be done prior to or in parallel with GIS development.

The main area where a combined, sustained effort to reach out and use GIS both toanalyse data and visualize stakeholder viewpoints – and is likely to pay off in termsof sustained benefits – is in projects addressing participatory area management.These could include CBNRM and participatory forest management, protected areaplanning and management projects, watershed-based and area-focused ruraldevelopment programmes, as well as educational planning and management.

AREAS FOR FURTHER DEVELOPMENT

P-GIS as a process tool

While GIS development should not be treated as an end in itself, P-GIS and itsproducts have a number of potential applications as a process tool in researchplanning and planning contexts:

• provision of a common frame of reference to enable more consistentinterpretation between stakeholders

• giving participatory planning authentic spatial references• driving discussions amongst different groups and between communities and

planners• assistance in providing visual aids to involve non-literate people in the planning

process• enabling early identification of issues requiring conflict resolution• provision of a visual and analytical tool in developing appropriate plans and

frameworks for an area management• strengthening stakeholder ownership of projects and project outputs• raising issues of access to information validates local information and can support

advocacy work• linking local and district level information, and encouraging planners and officials

to recognize and respond to community needs and aspirations.

Principal applications and further development of P-GIS

There is clear scope for application and further development of P-GIS in a numberof areas:


• planning and monitoring for locally based CBNRM and common propertyresource management in a number of areas, including developing participatorymanagement plans for collaborative forestry, rangelands and wildlife, wetlandsand aquatic resources;

• watershed or protected area management, especially where the objectives includeboth socio-economic development and environment/NR management;

• decision-making and planning for agricultural services and extension support –analysis of recommendation domains to enable targeting, and tailoring supportaccording to the characteristics of farming systems and local agro-ecological andmarket conditions;

• planning and monitoring for area-based rural and urban developmentprogrammes, not confined to NR sectors but extending to health, educational andbasic infrastructure planning.

In addition, there are potential uses of P-GIS in country strategy or sector-levelplanning applications which donors and governments could take up in the projectand programme design process. This, however, requires the existence of an adequateGIS database reflecting key stakeholder perspectives and knowledge and overalldiversity over the target area.

Research needs

Although stand-alone methodological research into P-GIS is not recommended, itssuccesses and weaknesses, and the lessons arising from ongoing and futureapplications can be monitored and drawn together.

A number of methodological and communications aspects would benefit fromfurther investigation in the context of ongoing applications and further NR research:

• effective use of perceptual information and local knowledge in scaling up for themanagement of wider areas, for example, in catering for the needs of differentfarmer and pastoralist groups and other resource users across a watershed ordistrict;

• use of GIS and GIS maps as effective visual aids, alongside other methods, and thefurther development of practical guidelines;

• application of P-GIS methods to the identification, management and resolutionof natural resource and planning conflicts;

• GIS for area-based projects taking a livelihoods approach, and as a managementtool to monitor livelihood change and development;

• role of GIS as a monitoring and evaluation tool for area-based developmentprogrammes, especially in a multi-agency, multi-stakeholder context.


P-GIS and the livelihoods approach

There is particular potential to explore the usefulness of P-GIS approaches inoperationalizing the sustainable livelihoods (SL) approach to development andpoverty eradication, which espouses the principles of people-centred stakeholderparticipatory development. The spatial perspective which GIS provides can assist inprioritizing, targeting and tracking the impact of interventions designed to improvelivelihoods and reduce poverty by:

• visualizing and mapping of capital asset availability over geographical space andchange over time;

• enabling an understanding of how vulnerability factors, socio-historical processes,and the effective reach of policies and service delivery institutions havedifferential impacts on different areas;

• using GIS as an overall cross-sectoral monitoring tool for progress in promotingSL and poverty eradication, incorporating a variety of indicators, such asincomes, food security measures, access to land, water supply and sanitaryconditions, access to basic services, animal and human health, employment,market development and trade flows (see Case Study 3).

Effective application of GIS to target areas, or even whole countries or regions, canhelp in monitoring and visualizing change and, thereby, measuring progress. Chartingthe geographical impact of the livelihoods approach requires the development ofspatial indicators of livelihood diversity, vulnerability, and access to livelihood assets(land, natural resources, infrastructure, markets, social institutions, skills and humanresources etc.), in addition to conventional poverty indicators. How far participatorymethods are needed to derive this information, and how far it is susceptible tospatial, visual representation will vary from place to place and according to scale.

Notwithstanding the methodological work that will be involved in making successfuldevelopmental and participatory applications, GIS undoubtedly provides a powerfultool for agencies committed to making a difference for the poor, and to strengtheningdialogue in planning and environmental management.

CONCLUSIONS

In setting out the lessons of recent participatory applications of GIS, and inexamining how the technology can be applied successfully by responsibledevelopment practitioners, some important general conclusions can be maderegarding the basics of good practice with P-GIS.

• Just as spatial analysis should not ignore society, social science should not ignorephysical space. In applying GIS, social science, natural science and spatial analysisshould be planned and implemented in co-ordination, and an interdisciplinaryapproach encouraged.

• In making effective applications, the development of the participatory process, atcommunity or institutional levels, is at least as important as the development ofGIS as a tool. The values of good participatory practice should be applied, withregard to ownership, expectations, iterative validation, communication, feedback


and sustainability. Institutional aspects of the use of both GIS and participatorymethods are critical.

• If the use of GIS is cost-effective, P-GIS adds considerable value at little extracost. However, adequate time, resources and expertise will be needed forparticipatory data collection (this should not cost more than a PRA study withoutGIS); and the iterative development of maps, plans and conclusions withstakeholders need to be budgeted.

• A P-GIS approach has considerable potential in prioritizing and monitoring forSL and poverty elimination. There is a case for well-planned regional investmentsin systems for planning and monitoring rural (or urban) development andprogress in poverty reduction in priority areas, using GIS and linked toprogrammes to promote accountable effective delivery by local government andsectoral planning agencies.

Finally, the risks of relying on GIS should be recalled:

• Once information finds its way into maps and computerized databases, there is atendency for users to regard it as immutable fact, even though it may be erroneousand partial, as for example, a single stakeholder’s perception of where a boundaryought to lie.

• GIS technology may be treated as an objective tool for scientists and planners todecree top-down, once and for all solutions to development problems, behindclosed doors, without input from primary stakeholders.

• Unless an integrated interdisciplinary approach is adopted from the start inrelation to data collection and analysis, the usefulness of GIS will be limited inaddressing real world development problems.

CASE STUDIES


Case Study 1: Use of GIS for coastal environmental planning and

management in north-eastern Brazil

Land use in the project area is complex and dominated by a small number of major landowners anddevelopers. The area is a mosaic of major tourist investment projects; private nature reserves,including Atlantic forest fragments, wetlands and dune systems; extensive farmland (coconuts andcattle); urban areas; loteamentos (land sold for housing, holiday homes and small-scale touristdevelopment); pine and eucalyptus plantations (for pulp, cellulose and sawn-timber markets); andlimited small-scale farming.

The project GIS, originally intended as a centralized planning tool, is developing an up-to-date,database of the project area for thematic mapping and analysis of spatial data. A participatorysocio-environmental survey has been made in the pilot area; and a series of local stakeholders’consultations has been held, with the aim of developing a consensus-based planning programmefor environmental protection and socio-economic development. There are good opportunities toincorporate the outcomes of social and participatory enquiry in the GIS and thematic maps, andto apply the GIS in support of stakeholder participation in planning in a number of related areas.


Overall project planning and co-ordination

• Different state institutions and other sources have contributed data to the GIS and the integrateddatabase will be made available for stakeholders’ use as a focus for institutional collaboration.

• Development of maps using GIS has promoted an interdisciplinary approach, spatial thinking bysocial scientists, and more user-oriented approaches by physical planners. Social developmentcritique of early map products enabled the development of more accessible, user-friendlyversions.

GIS as a tool to assist survey planning

• The GIS enabled the reconciliation of census data with administrative and planning units withinthe area, and rural settlements identified and described by participatory fieldwork were geo-referenced and incorporated in the GIS to develop a population sampling frame for surveys.

• Aerial photographs and participatory sketch maps were used to develop maps to assist insampling in towns, and for use in urban environmental management.

Integration and management of diverse datasets

• Socio-economic datasets including human settlements, populations, roads, land ownership andbasic services were added to baseline (1:250 000) topographical maps.

• Ranked information generated by PRA and sample surveys, including livelihood activities, andpoverty indicators (income levels, capital asset, environmental health, service access and literacydata) was incorporated and represented visually.

• Selected datasets were then overlaid to produce integrated 1:100 000 scale maps, e.g. includingsettlements, population, principal livelihoods activities with land ownership, and land cover.

Maps to promote stakeholder participation

• The GIS enables organization of information at different scales for display and feedback todecision-makers and stakeholders at different levels; this is particularly useful to promoteunderstanding amongst state agencies and municipalities, and at community level.

Promising applications of a participatory GIS approach include:

• comparison of poverty and environmental indicators across and between different zones andlocations within the project area

• identifying potential extractive reserves (for fishing, shellfish and plant fibre resources) anddeveloping local management plans with users and landowners

• small-scale ecotourism development and planning with local business people

• refining and improving larger-scale maps (1:25 000) of local river catchments and villageperipheries through discussion with local people – and making them available to communitygroups as a visual aid for local development and environment projects.

Principal lessons learned to date include:

• value of using GIS to integrate socio-economic and spatial perspectives from the beginning

• importance of fostering an enabling institutional environment if a centrally operated GIS is tobecome an effective tool for participatory planning

• the need for longer time frames to allow for the development of appropriate institutionalarrangements and the GIS database itself.

Case Study 1 cont.


Case Study 2: Combining scientific and indigenous knowledge of soils using

GIS in Uganda and Tanzania

One of the project’s main objectives was to develop methodologies for comparing and combiningscientific and indigenous knowledge of soil and land resources using GIS as an integration domain.GIS was used for spatial analysis of the soil data, collected through scientific soil surveys, and ofindigenous knowledge of soils and land resources, explored using social-anthropological andparticipatory rural appraisal techniques. GIS proved to be a useful tool for integrating and combiningboth sets of information although methodological and practical issues, such as data collection and itssequencing and continuous availability of GPS/GIS equipment during fieldwork, need carefulconsideration. The critical link between the work on local knowledge and the scientific survey wasgeo-referencing the observations. However, a proper functioning GPS was not available throughoutthe fieldwork and, therefore, not all information gathered could be successfully integrated andanalysed within the GIS.

Furthermore, exploring indigenous knowledge is a delicate process as the use of GIS tends to focusmainly on spatial analysis, e.g. through use of perceptual maps. This research illustrates clearly thatthere is much benefit in a detailed and carefully sequenced process of exploring local people’sknowledge, drawing on different tools and contexts. It was only through the combination of groupdiscussions, individual household interviews, transect walks and participatory mapping that an in-depth understanding of local soil classifications and the differences in cognitive processes began to bedeveloped.

This information is very relevant for supporting and improving the analysis done within the GIS. Inaddition, existing GIS techniques can be used creatively to integrate different datasets, such as hot-linking text files to perceptual and scientific maps, and overlays of geo-referenced householdinformation and local soil categories with scientific land resources maps.

Additional lessons learned are, in brief:

• close integration is important – detailed investigation of farmers’ categories and concepts withphysical observations and use of base maps, photographs and GPS must all be integrated;

• consistency in the participating groups is needed – the composition both of the farmers’ groupsinvolved in the study and of the research teams should be consistent;

• phasing is important – detailed indigenous knowledge studies, to enable in-depth understandingof categories and content, should precede serious attempts to consolidate this knowledge throughfarmers’ mapping;

• research involved an important learning process for all participants – it is important that theresearch team has training to develop a thorough understanding of GIS and its potential beforestarting the fieldwork.

Case Study 3: Participatory land use planning in two villages in Kumasi, Ghana

One of the issues arising from previous research in the Kumasi region is the lack of participation bylocal people in land use planning. The final phase of the Kumasi Natural Resources ManagementResearch Project used the findings of previous research and mapping information in the KUMINFOGIS to develop two pilot projects in the villages of Swedru and Aburaso.

The objective was to enhance land use planning by:

• increasing participation of local stakeholders

• promoting the sustainable use of natural resources in the planning process

• considering environmental issues in the planning process

• taking into account inter-village, watershed and regional issues at village level planning

• developing linkages between the village and district and regional planners and other professionalsso that the villages may benefit from appropriate professional inputs.

Activities undertaken with chiefs and elders to discuss the planning process:

• geo-referenced copies of aerial imagery were printed

• meetings held with chief and elders, women, men, youth, poorer groups to map the naturalresources and other features on the images

CONSIDERATIONS IN CHOICE OF P-GIS HARDWARE AND

SOFTWARE

Guidance is provided on the technical requirements for spatial analysis of complexdatasets, participatory mapping and maps to communicate geo-spatial issueseffectively to stakeholders.

• It is often impracticable to have complicated software and hardware at thecommunity level and there is great danger in placing the technology in an ill-prepared local institution.

• Available software ranges from expensive fully comprehensive systems suited toserious GIS laboratories to desktop display GIS with limited analysis andintegration functions, some of which are free. Costs of the software also varyconsiderably and costs of maintaining an upgrade strategy can be prohibitive,especially after project support has finished.

• The choice of software should be made at the planning or inception stage of aproject – it should not be considered ad hoc as the project advances. The software


Case Study 3 cont.

• the maps were used to identify problems

• meetings with district planners and other professionals to invite their involvement

• ideas were pooled to develop a village plan

• modest initial funds provided, and action research facilitated finding, the necessary technicalexperts

• final maps were digitized into KUMINFO.

Reactions to the use of the images and maps:

• many local people could interpret the images and could locate their own homes, farms, watersources etc., and there was considerable interest in identifying other features such as ridges, valleysstreams and developments.

• the maps encouraged common perception of the same resources and helped to reduce conflictsdue to misunderstandings, for example, over boundaries.

• because of overhanging or dense vegetation that sometimes obscured the paths and small streams,the images proved more difficult to interpret in the less used areas.

Village level action was then prioritized and an agreed practical project identified to the satisfactionof all groups in each village – in Aburaso, the provision of a public hand-dug well and, in Swedru, theprotection of a stream by tree planting. Both communities formed project implementationcommittees. Experts from the local district water and sanitation team and the Land andEnvironmental Management Office became involved and both projects are now self-sustaining.District planners have shown interest in this approach, although with mixed reactions.

The research team concluded that the GIS-produced images enabled the villagers to consider planningactively for the natural resources around them for the first time. The process, supported by the GISmapping exercise, showed villagers that the loss and degradation due to development were notinevitable, and that interventions and the wishes of the groups normally excluded from decisions weretaken into account to identify beneficial small-scale interventions. Moreover, the GIS developed hasthe potential for similar applications more widely around Kumasi and to provide a tool for extendingand scaling up dialogue amongst local communities, chiefs, planners and local government officials.

should be adequate for the project, but should not be overbought either incomplexity or price.

• The version of GIS software and ancillary software (word processing,spreadsheets and databases) should not be changed during the life of a projectunless all stakeholders agree to change – and adequate funds are available for auniversal change.

• Most software now runs using PC technology and can be used on a laptopcomputer in the field – both may be considered for data collection – and there areportable data-loggers for collection of GPS data and attribute data. However,these options are of little long-term use when the provision of electricity is non-existent or intermittent.

The choice of GIS software and hardware should always be considered in relation toseveral factors:

• analytical needs of the project• institutional capability, in human, physical and sustainable resources• costs of the hardware and software (see table)• costs of training.

Use and acquisition of software

In general, several principles can be applied in the use and acquisition of GISsoftware and hardware.

1. To answer the needs of a national or regional institute where data, especially basebiophysical data, can be shared between many users and several projects, a top-end system should be considered. This will require a commitment to continuetraining and technical support on the part of donors as well as a considerableinstitutional commitment on behalf of the recipient. Such institutionaldevelopment support will almost certainly be beyond the scope of a limitedresearch project.

2. Where a top-end GIS already exists and the local institutional framework allows,projects should seek to use the technical and physical support this might allow.While there may still be a commitment to project provision of a ‘field’ or‘project’ GIS, the use of existing resources, even on an ‘agreement’ basis cuts thecost of acquiring base data and hardware (e.g. large format plotter and digitizer).

3. For research projects, the software and hardware will usually be PC- or laptop-based. It is important, however, to provide adequate means to input and outputdata. Unless digitizing is required in the field, local and project support servicesshould be sought before a digitizer is bought. Such equipment is probablyunjustified within a research project time frame unless it can be justified for localinstitutional strengthening. A similar argument applies to a large format plotter(A0). Smaller format (A3) printers give excellent output with most GIS software.

4. How the outputs are to be used within the participatory context should beconsidered at the planning stage. If paper ‘products’ are to be used in the field, auseful piece of hardware is a laminator – this not only provides protection but


also enables the product to be written on with a non-permanent marker (e.g.chinagraph) during discussions.

5. The ownership of data generated by project should be considered. Where acommunity has access to even limited hardware, access to a free GIS viewer canbe provided so that data can be used locally.

GLOSSARY OF COMMON TERMS AND ACRONYMS USED

attribute non-geographic data associated with a spatial element (point,line, area) in a GIS

base maps topographic and thematic maps that form a background tothe spatial analysis in a GIS

CBNRM community-based natural resource management

DBMS database management system – a set of computer programsfor organizing information in a database; typically containsroutines for input, verification, storage, retrieval andcombination

fuzzy boundaries perceptions on boundaries of, for example, land ownershipmight differ among the different stakeholders involved orboundaries of soil categories might be based onobservations and estimations during scientific soil surveys


Top end

Middle range

Bottom range

Free

Examples

Intergraph MGEArcInfo SPANS

ARCVIEW Mapinfo TacticianILWIS

IDRISIMapmakerMaptitude Atlas

Arc-explorerMapinfo viewer

Type

Workstation,e.g. unix, NT

PCLaptop

PCLaptop

PCLaptop

A0 plotter

A3 printer

A4 printer –inkjet

Cost of softwareand extensions

£15 000+

£5 000

n/a

0

SOFTWARE

Costs – basic

£10 000+

£1 500

<£1 000

0

Costs

£4 000

£2 000–£3 000

£2,000–£3 000

£1 000

£4 000

£400

£100

Cost of software and hardware

HARDWARE

PERIPHERAL HARDWARE

fuzzy logic fuzzy logic or fuzzy set theory allows individual pieces ofdata to be members of different overlapping sets. It usespossibility instead of probability as a statistical technique;the result is a graduated boundary rather than a crisp linearone

GIS geographical information system/s – a computer-based toolfor mapping and analysing data related to space; GIS allowsfor the integration of disparate layers of information into acommon spatial database, whether the data are related to anexplicit reference (latitude/longitude, grid) or to an implicitreference (e.g. census tract name, road, environmentalfeature)

GPS global positioning system – a device for determininggeographical location anywhere on the earth’s surface, usingelectronic receivers to obtain latitude, longitude and altitudedata from satellites in geo-stationary earth orbits. GPS canbe especially useful when map coverage is limited and/or outof date. It also enables addition of specific informationrelevant to particular projects such as the distribution andlocation of schools, water wells, soil types and emergingsettlements. GPS is used increasingly in data collectionexercises as it is a cheap and user-friendly. Portable GPSconsists of a hand-held receiver used to receive signals fromsatellites to reference the points at which specific features arelocated. This information is displayed and stored in thehand-set and can be downloaded into a computer system.Although GPS is cheap and easy to use in the field, its mainlimitation is accuracy – a result of in-built errors of 10–30metres due to interference from US Defence satellites. Byusing a local base station on well-located objects, such as aroad junction or established buildings, it is possible to usedifferential GPS measures to improve accuracy and correctfor these errors. The use of a base station involves highercosts and skills required to set it up, so it is worthwhile toconsider the level of accuracy required according to thenature of the data to be collected and the objectives of theproject

ground-truthing verification of remotely sensed data by inspection andcorrelation in the field

hot-linking hyper link between related pieces of information; a hot-linkallows non mapable data to be referenced to a geographicalfeature (point, line or polygon) and then retrieved anddisplayed by the click of a button. Almost any informationcan be handled this way and can include text files, charts,tables, pictures, photographs and video clips

livelihood comprises the capability, assets and activities required for ameans of living

livelihoods starting point is an holistic analysis of people’s livelihoodsacross sectors, areas and social groups, recognizing the


multiple and overlapping realities, dynamic linkages andchanges over time

LPK local people’s knowledge – includes the complex ofpractices and decisions made by local people and shaped byand interlinked with technical, cultural, political and socialknowledge. Local knowledge is based on experiences passedfrom one generation to the next, but nevertheless it changes,adapts and assimilates new ideas. It can be quite location-specific and may vary among individuals according to age,gender, socio-economic status, ethnicity and area ofresidence

meta data data about available datasets and sources

network diagram represent the multiple inter-relationships between e.g. themany components of farming, livelihoods systems orinternal and external linkages

NGO non-governmental organization

NR natural resources

overlay (noun) data plane containing a related set of geographic datain digital form; (verb) process of stacking individual digitalrepresentations of various spatial data so that each positionin the area can be analysed in terms of these data

P-GIS participatory GIS – integration of local knowledge andstakeholders perspectives in the GIS; stakeholders shouldalso have access to GIS databases and GIS products, and beable to apply GIS and GIS products to developmentplanning, resource management and advocacy

participatory mapping ideally participants’ free drawn maps showing particularfeatures of relevance to their livelihoods, e.g. naturalresources, their village, social resources (aspects of socialrelations and household distribution)

participatory planning involvement of a community in planning the use of acommon resource, identifying common goals towards themanagement of the resource and implementing a plan toachieve those goals

PRA participatory rural appraisal – empowerment-orienteddevelopment appraisal that is initiated by an externalmultidisciplinary team, using qualitative research methods,in order to help a local community conduct an efficientassessment of its own situation, including problems andpotential

photo/image maps aerial photographs, aerial images or satellite images whichhave been geographically corrected and probably overlainwith key topographic data to provide a map output

pixel smallest unit of information in a raster (grid cell) map or ina scanned image


ranking exercises evaluation of particular properties according to the criteriaof the participants, e.g. farmers’ preference for maizevarieties. Farmers’ own criteria are listed in the left-handcolumn, then the attributes of the different varietiesdiscussed. Local materials such as stones and beans can beused for scoring, providing a quantitative expression ofpreferences

remote sensing acquisition of data by a means of a remote sensor; usuallyrefers to satellite or airborne acquisition of raster imageryor photographs

seasonal calendar a graphical device with the months of the year on thehorizontal axis and a set of activities, such as farmingoperations or livelihood and employment activities, on thevertical key. The activities and their timing are derived fromwider group discussion and the technique is useful to obtainan understanding of farming and livelihood systems in acommunity. It can illustrate patterns of labour and incomeover the year of different social groups within thecommunity (gender, age, wealth and ethnicity)

sorting interviews obtaining an understanding of local classifications and theirinterrelationships. Concepts of, e.g. soil types or trees, aresorted and grouped by local participants into categories,individually identifying common features and contrasts inthe context of the wider language and cultural system

stakeholders persons, groups or institutions with interests in a projectprogramme. Primary stakeholders are those ultimatelyaffected, either positively (beneficiaries) or negatively (thoseinvoluntarily resettled). Secondary stakeholders are theintermediaries in the aid delivery process. Stakeholdersinclude both winners and losers and those involved orexcluded from decision-making processes. Key stakeholdersare those who can significantly influence, or are important tothe success of the project

thematic map map containing data related to a specific theme, e.g. soils,population density, land suitability etc.

topographic map map showing the surface features of the earth’s surface, e.g.roads, contours, rivers etc.

transect walks systematic walks to explore local practices; researchersobserve, ask questions and listen, and farmers talk anddescribe their land, natural resources and farming systems(how and why they do things)

UPS uninterruptible power supply – a device for stabilizing theelectrical input to computer hardware, protecting againstpower surges and storing energy for protection againstpower cuts

workstation a mini computer or high level PC, often networked to othercomputers


INSTITUTIONAL CONTACTS

Photomaps

Nepal photo-map work is being considered for expansion to most of the communityforest area of Nepal, and is being taken up by other bilateral projects under the co-ordination of DFID’s Nepal-UK Community Forestry Project.

Key contacts:

Mr Peter Neil, Project Co-ordinating OfficerNepal-UK Community Forestry ProjectPO Box 106, Kathmandu, NepalDr Richard A. MatherForest Products Research Centre, Buckinghamshire University CollegeQueen Alexandra RoadHigh Wycombe, HP11 2JZ, UKTel. +44 (0) 1494 522141 Ext. 3214 Fax. +44 (0) 1494 605051e-mail: [email protected]

P-GIS applications for natural resource management in South Africa

Key contact:

Steve CinderbyStockholm Environment Institute at York (Deputy Director)Box 373, University of YorkYork YO10 5YW, UKTel: +44 1904 432994 Fax: +44 1904 432898Website: www.seiy.org

FURTHER READING

ABBOT, J., CHAMBERS, R., DUNN, C., HARRIS, T., MERODE, DE E., PORTER,G., TOWNSEND, J. and WEINER, D. (1998) Participatory GIS: opportunity oroxymoron? PLA Notes, No. 33. London: International Institute for Environment andDevelopment.

ALDENDORFER, M. and MASCHNER, D.G. (1996) Anthropology, Space and

Geographical Information Systems, Spatial Information Systems Series. NewYork/Oxford: Oxford University Press.

BURROUGH, P.A. and MCDONNELL (1998) Principles of Geographical Information

Systems, Spatial Information Systems Series. New York/Oxford: Oxford UniversityPress.

CHAMBERS, R., PACEY, A. and THRUPP, L.A. (eds) (1989) Farmer First: Farmers’

Innovation and Agricultural Research. London: Intermediate Technology PublicationsLtd.

CINDERBY, S. (1999) Geographic Information Systems (GIS) for participation: thefuture of environmental GIS? International Journal Environmental Pollution, 11(3):1–12.


DAPLYN, P., CROPLEY, J. and TREAGUST, S. (1994) The Use of GIS in Socio-

economic Studies. NRI Socio-economic Series 4. Chatham, UK: Natural ResourcesInstitute.

FERBEE, L. (1989) A folk expert system: soil classification in the Colca Valley, Peru.Anthropological Quarterly, 62(2): 83–102.

FLOWERDEW, R. (1994) Spatial integration. pp. 337–359. In: Geographical

Information Systems, Principles and Applications. Maguire, D.J., Goodchild, M.F. andRhind, D.W. (eds). London: Longman Scientific and Technical and New York: JohnWiley.

HARRIS, T.M., WEINER, D., WARNER, T.A. and LEVIN, R. (1995) Pursuingsocial goals through participatory GIS: Redressing South Africa’s historical politicalideology. pp. 196–222. In: Ground Truth: The Social Implications of Geographic

Information Systems. Pickles, J. (ed.). New York: The Guildford Press.

HARRIS, T.M. and WEINER. D. (1997) Empowerment, marginalization andcommunity integrated GIS. Cartography and Geographic Information Systems, 25: 2(April).

HUTCHINSON C.F. and TOLEDANO (1993) Guidelines for demonstrating GISbased on participatory development. International Journal of GIS, 7(5): 453–461.

LAWAS, M.C.M. (1997) The Resource User’s Knowledge, the Neglected Input in Land

Resource Management. The Case of the Kankanaey Farmers in Bengeut, Philippines, ITC

Publications, No. 52. Enschede, the Netherlands: International Institute forAerospace Survey and Earth Sciences.

MATHER, R.A. (2000) Using photomaps to support participatory processes ofcommunity forestry in the Middle Hills of Nepal. Mountain Research and

Development, 20(2): 154–161.

MATHER, R.A., DE BOER M., GURUNG, M. and ROCHE, N. (1998) Aerialphotographs and photo-maps for community forestry. pp. 13–22. In: Rural

Development Forestry Network, ODI Network Paper 23, London: OverseasDevelopment Institute.

PICKLES, J. (ed.) (1995) Ground Truth: The Social Implications of Geographic

Information Systems. New York: The Guildford Press.

RUNDSTROM, R.A. (1995) GIS, indigenous people and epistemological diversity.Cartography and GIS, 22: 45–57.

TABOR, J.A. and HUTCHINSON, C.F. (1994) Using indigenous knowledge, remotesensing and GIS for sustainable development. Indigenous Knowledge and Development

Monitor, 2(1).

VEREGIN, H. (1995) Computer innovation and adoption in geography. pp. 88–112.In: Ground Truth: The Social Implications of Geographic Information Systems. Pickles, J.(ed.). New York: The Guildford Press.


WARBURTON, H. and MARTIN, A. (1999) Local people’s knowledge in naturalresources research. Socio-economic Methodologies for Natural Resources Research. Best

Practice Guidelines. Chatham, UK: Natural Resources Institute.

WEINER, D., WARNER, T.A., HARRIS, T.M. and LEVIN, R.A. (1995) ApartheidRepresentations in a digital landscape: GIS, remote sensing and local knowledge inKierpersol, South Africa. Cartography and Geographic Information Systems, 22(1):30–44.

WOODFINE, A.C. (1995) Geographical Information Systems as Appropriate Technology

in Developing Countries. Chatham, UK: Natural Resources Institute.

Website:www.geo.vuw.ac.nz/geography/projects/migis Case study of a P-GIS projectin China, highlighting the research process, methodology and lessons learnt.


gis and participatory approaches in natural resources research

Documents